Hi Folks, hope you all had a great Halloween!
Getting back to this, not much left to do except finalize the boards. Will be doing that over the coming days.
Getting back to this, not much left to do except finalize the boards. Will be doing that over the coming days.
Here's the final schematic. Gerbers coming soon.
The big change is the buffer. It is now a push-pull buffer and I am finally satisfied with how it sounds compared to the no buffer version. It uses the n-channel J113 and a p-channel J176 with source resistors to set the bias current. With this J113-J176 buffer, the circuit is now truly complete and can drive anything. I'm using it to drive a Lateral FET MOFO, which is somewhat current hungry and it drives it fine. I want to share this lateral MOFO with all of you as well--it sounds so good.
The PCB for this version is almost done and is the full size one with switching and volume control, but I'm planning to post another smaller one that can be put closer to the power devices in an amp or be used to update the older SCG PCB.
The big change is the buffer. It is now a push-pull buffer and I am finally satisfied with how it sounds compared to the no buffer version. It uses the n-channel J113 and a p-channel J176 with source resistors to set the bias current. With this J113-J176 buffer, the circuit is now truly complete and can drive anything. I'm using it to drive a Lateral FET MOFO, which is somewhat current hungry and it drives it fine. I want to share this lateral MOFO with all of you as well--it sounds so good.
The PCB for this version is almost done and is the full size one with switching and volume control, but I'm planning to post another smaller one that can be put closer to the power devices in an amp or be used to update the older SCG PCB.
@ra7
Have you seen @Tubelab_com last update on the load for this kind of configuration?
I tried it and it works really well, better than CCS IMHO.
Have you seen @Tubelab_com last update on the load for this kind of configuration?
I tried it and it works really well, better than CCS IMHO.
Because J111 has a wider Vgs range, providing more headroom, and that works better in this design. J175 sounds marginally better than J176 and J74, I don’t know why. One is free to experiment and can use others. On larger swings, the J175-J111 combo keeps the inherent harmonic profile intact, whereas with others you start seeing higher levels of higher order harmonics and it is heard (to me at least) as a strained quality in the sound.
I did see that. But you essentially get the same with the circuit here with fewer parts. R204 communicates the signal at the drain of Q202 to the gate of Q203, thereby adjusting the AC current and keeping the impedance to Q202 high and output impedance low. In George’s version C2 plays that role, similar to the gyrator circuit.@ra7
Have you seen @Tubelab_com last update on the load for this kind of configuration?
I tried it and it works really well, better than CCS IMHO.
I tried the option of taking the drain feedback from the source pin of the CCS FET, but preferred the more pure Schade version. In the version I posted above, you can even take the feedback from the output buffer’s output node. But I again prefer the local feedback in terms of sound. You lose a bit of magic with the global feedback. There may be other ways to incorporate that. I have a few ideas based on this topology.
18 db. Slightly less than previous versions because of the loss in the input and output buffers.
Did not know that. But then - why have one for the J113 in the current source?
Will never complain about fewer parts.
Take care, Rahul
Will never complain about fewer parts.
Take care, Rahul
When Michelangelo worked on the statue of David, he started with a large block of silicon and then removed any silicon that was not David.
Rahul is doing the same here! Go, go, Rahulangelo!
Rahul is doing the same here! Go, go, Rahulangelo!
Aww, you guys are makin' me blush. Yes, that is what I was trying to do. Why keep a part if it is not serving any purpose.
Working on resistor footprints. KiCAD doesn't have a big pad small resistor footprint, so I'm creating one. I had a bit of trouble with the small sized resistors, especially after basterdizing the one PCB for several experiments. For the final, I'd like to have bigger pad sizes for even the small resistors.
Working on resistor footprints. KiCAD doesn't have a big pad small resistor footprint, so I'm creating one. I had a bit of trouble with the small sized resistors, especially after basterdizing the one PCB for several experiments. For the final, I'd like to have bigger pad sizes for even the small resistors.
When I order boards, I always make sure to get those nice pads and plated through-holes.
My soldering is pretty good and I really get nice joint that ensures beyond necessary conductivity.
Alas, when I need to replace any part for any reason, those nice plated through-holes become my worst nightmares, as they are quite hard to desolder.
Hi Rahul, I'm very curious on this, I'm a big fan for laterals (and iron...)
For what the SCG pre is concerned, what are the matching criteria for JFETs? Idss, VP? Would you be so kind to show the parameters of the JFETs you are using in the real circuit? (as I've got some old stock Toshibas that I would use)
Thanks
Guido